Munich Personal RePEc Archive
Growth and Poverty Reduction in West Africa: A Brief Overview
Wodon, Quentin
World Bank
January 2007
Online at https://mpra.ub.uni-muenchen.de/11086/
MPRA Paper No. 11086, posted 16 Oct 2008 01:55 UTC
Growth and Poverty Reduction in West Africa—
A Brief Overview
By Quentin Wodon
This volume provides a set of six case studies from West Africa in order to contribute to an assessment of the benefits of growth (or the cost of a lack of growth) for poverty reduction in those countries. The first part of the volume describes the experience of two countries (Ghana and Senegal) that achieved high levels of growth in the 1990s, and that also experienced important reduc- tions in poverty, even though growth was not strictly pro-poor. The second part of the volume describes the experience of two other countries (Burkina Faso and Cape Verde) that also achieved high levels of growth in the 1990s, but where due to data and methodological issues, there was an initial perception that growth did not lead to much poverty reduction. The more detailed analysis of poverty presented here suggests however that these two countries did witness a sharp reduction in their population share in poverty, as would have been expected given their growth record. Finally, in the third part of the volume, it is argued that a lack of growth in the 1990s in the last two countries under review (Guinea-Bissau and Nigeria) has been a key reason for their persis- tently high levels of poverty. Overall, the case studies in this volume make a strong case for the positive impact of growth for the reduction in poverty in West Africa, but they also point to the need to pay close attention to changes in inequality as such changes have limited the gains from growth for the poor in several of the countries considered here.
P
overty as traditionally defined using household income or consumption data is a function of both the level of mean income or consumption in a population, and the inequality in the distribution of these resources. Hence, both growth (a change in the mean) and changes in inequality can lead to changes in poverty. There is consensus that broad economic growth is important for poverty reduction. Economic growth typically1
leads to higher household incomes, which in turn enable households to increase their con- sumption and thereby reduce their probability of being poor. However, there are also con- cerns that growth often may not be enough, and that policies to reduce inequality also are important for poverty reduction. These concerns were echoed in the latest report of the World Bank’s Independent Evaluation Group on development effectiveness.1They are also echoed in discussions within Sub-Saharan African countries. For example, several Poverty Reduction Strategies adopted by governments in West Africa have discussed the apparent lack of a strong enough link between economic growth and poverty reduction.
The authors of the papers in this volume are certainly convinced that a reduction in inequality is important for the reduction of monetary poverty, and the work presented in this volume does suggest that inequality is actually worsening in many of the countries con- sidered here. At the same time, the authors also argue that without strong economic growth, it will be difficult to achieve sustainable poverty reduction and reach the Millen- nium Development Goal target of halving extreme poverty by 2015. Especially for African countries which tend to be poorer, the long-term gains in terms of poverty reduction that can be obtained from growth tend to be larger than those that can be achieved through the reduction in inequality, simply because there is often not that much initially to redistrib- ute. This does not mean that policies towards inequality reduction should not be consid- ered. However, it does imply that growth should remain central in the Poverty Reduction Strategies prepared and implemented by developing countries.
Until recently, good household survey data that were comparable over time were lacking in many African countries, including in West Africa. This made it difficult to properly mea- sure changes in poverty over time, and assess the links between growth, changes in inequality, and changes in poverty. Today however, in part thanks to the monitoring and evaluation com- ponents of the countries’ poverty reduction strategies, the availability of good survey data is becoming more widespread. In the context of a larger effort at the World Bank to measure changes in poverty over time in Sub-Saharan African countries, the objective of this volume is to document the trends in growth and poverty reduction in six West African countries (Burkina Faso, Cape Verde, Ghana, Guinea-Bissau, Nigeria, and Senegal). In four of the six countries, the focus is on poverty measures obtained by comparing results from two house- hold surveys over time. In the last two countries, because only one household survey was avail- able for the work presented here, the analysis focuses somewhat more on growth itself (or the lack thereof) than on poverty reduction, although poverty measures are estimated and esti- mations of their trend over time are provided using simple simulation techniques. These somewhat naive simulation techniques certainly have important limits, but they are never- theless instructive to give a rough idea of the potential impact of a lack of growth on poverty.
The objective of this introductory chapter is to provide a quick overview of the key find- ings from the six case studies, as well as a brief conclusion on the lessons learned from this work.
Overview of the Case Studies
The first part of the volume is devoted to Ghana and Senegal, two countries with high rates of economic growth in the 1990s and substantial poverty reduction, even though growth in each of these two countries could not be considered as strictly pro-poor.
2 World Bank Working Paper
1. See http://www.worldbank.org/ieg/arde2006?intcmp=5287207.
Chapter 2 documents the record of poverty reduction over the 1990s in Ghana, when macroeconomic performance remained good, although less consistent than in the years directly following the reforms that started in 1983. The analysis is based on comparable household surveys conducted at the beginning and end of the decade. At the national level poverty fell, and the evidence for this seems quite robust. Non-monetary indicators of well-being available from the surveys also show improvements over time, except for use of health care facilities. Yet, poverty reduction was concentrated in specific areas (including Accra and the rural forest region) and particular household activities (notably export oriented sectors and commerce). Households in other localities and working in other activities experienced little poverty reduction, with some evidence of increasing depth or severity of poverty in some locations (especially in the northern savannah). Thus, despite the reduction in the overall poverty measures, this paper highlights the limited benefits accruing to many of the poorest groups in Ghana over the 1990s and the increased differential between localities that emerged over this period. This could be a consequence of the country’s less consistent macroeconomic performance over the period as compared to the post-reforms 1980s, but it also raises questions about the overall policy stance and the extent to which it focused enough on poorer, more remote regions and on non-export agriculture.
Chapter 3 on Senegal starts with the fact that the devaluation of the CFA Franc in 1994 generated a public investment boom. The increase in public investment was made possible thanks to an improved budgetary situation related to the reduction in real terms of the public wage bill which had been too large for some time (the wage bill was reduced because nominal wages were not adjusted for the implicit loss in purchasing power that the deval- uation entailed). The rise in public investment was subsequently accompanied by a (smaller) increase in private investment due in part to the attractiveness of Senegal as a place to do business within West Africa, at least compared to other West African nations (this attractive- ness has recently been further enhanced with the crisis in Côte d’Ivoire). In turn, higher public, and to some extent private, investment led to higher growth rates and substantial poverty reduction, with the share of the population living in poverty declining from 67.9 to 57.1 percent. However, as was the case in Ghana, poverty in urban areas was reduced faster than in rural areas, as most of the investment benefited the manufacturing and services sectors. Also, a few years of poor rainfall in the second half of the 1990s coupled with an initial drop in the real prices of crops in the aftermath of the devaluation affected negatively rural incomes. As a result, while virtually all segments of the population (including the rural poor) benefited from improved standards of living in 2001 as compared to 1994, growth was not strictly speaking “pro-poor” because the growth in consumption per equivalent adult in the upper half of the distribution was larger than that observed among the poor.
The second part of the volume describes the experience of two other countries (Burkina Faso and Cape Verde) that also achieved high levels of growth in the 1990s, but where due to data and methodological issues, there was an initial perception that growth did not lead to much poverty reduction. The more detailed analysis of poverty presented here suggests however that these two countries did witness a sharp reduction in their population share in poverty, as would have been expected given their growth record.
Chapter 4 provides estimates of the trend in poverty and inequality in Burkina Faso.
In order to generate information about living standards, poverty, and inequality, the country implemented three Priority Surveys in 1994, 1998 and 2003. Throughout this period, efforts were made to improve the design and implementation of the surveys, from sampling and
questionnaire design, to data entry and quality control. As a result, the precision of the latest poverty and inequality estimates was improved, generating more accurate and nuanced information on living standards. The improvements in survey design increased the precision of the newest poverty estimates, but endangered the comparability with the poverty estimates produced by earlier surveys. As a result, the initial official poverty and inequality estimates using the three surveys, which suggested a lack of progress towards poverty reduction, were not based on comparable data sets. The authors in Chapter 4 have defined an alternative and comparable welfare indicator, constructed using the same methodology and based on a subset of consumption items that were identically recorded across the 1998 and 2003 surveys.
Based on this welfare aggregate, the poverty head count went down from about 54.6 per- cent in 1998 to 46.4 percent in 2003. The reduction in poverty was more pronounced in rural than in urban areas. In other words, a closer examination of the data suggests that the paradox of high growth without poverty reduction was actually due in large part to measurement errors.
Chapter 5 is devoted to Cape Verde, a country that shifted from a socialist to a capi- talistic model in the late 1980s. This shift enabled the population to benefit from rapid eco- nomic growth, but concerns have been expressed about a potential increase in inequality.
Two household surveys with consumption data implemented in 1988–89 and 2001–02 provide information that can be used to assess the impact on welfare of this policy shift.
Initial estimates based on these two surveys suggested that there had been an increase in poverty over time, but this was mainly due to the adoption of a relative measure of poverty and to comparability issues between the surveys. The task of assessing the trends in poverty and inequality was also made more difficult because the unit level data of the first survey are not available. For the period 1988–89, the only information at our disposal consists of a number of tables on the distribution of income in the original report prepared 15 years ago on that survey. This makes it necessary to estimate poverty and inequality using group data.
In Chapter 5, the authors use the Poverty module of SimSIP (a simulation tool from the
“Simulations for Social Indicators and Poverty” project) in order to obtain new poverty and inequality trends over time with group data. They find that despite an increase in inequality over time, and thereby an increase in relative measures of poverty, absolute poverty measures have been reduced dramatically thanks to rapid growth.
Finally, in the third part of the volume, it is argued that a lack of growth in the 1990s in the last two countries under review (Guinea-Bissau and Nigeria) has been a key reason for their persistently high levels of poverty. As for the reasons for the lack of growth in each of the two countries, they differ. Conflict was a major factor in Guinea-Bissau, and macro- economic instability played a key role in Nigeria.
In Chapter 6, it is argued that conflicts and political instability have been serious con- strains to growth in Guinea-Bissau. Of special concern was the civil war of 1998, which lasted 11 months and led to substantial loss of life as well as to a massive decrease in GDP per capita. Based on previous work on the economic cost of conflicts in Sub-Saharan Africa and using a technique to identify outliers in time series and correcting the series for such outliers, the authors estimates that GDP per capita today could have been more than 40 percent higher if there had not been a conflict in 1998. In turn, if one is willing to assume that GDP per capita is closely correlated to the level of consumption per capita of households, about one in three persons today in poverty might not have been poor without the negative impact of the conflict. While there are a number of assumptions involved in these estimations, they 4 World Bank Working Paper
do clearly suggest that conflicts can have a devastating impact on the well-being and poverty levels of a population.
The last Chapter is devoted to Nigeria. At the time at which the paper was written, the latest nationally representative survey for Nigeria was from 1996, and estimates suggested that two thirds of the population was poor in that year. The authors argue that this high level of poverty was due in large part to macroeconomic volatility that depressed private investment and growth. Using cross-sectional data for 87 countries, they show that real per-capita growth over the period 1980–94 was a function of productivity growth and investment rates, both of which were negatively effected by volatility (in terms of trade, real exchange rate, and public investments). When comparing Nigeria to high growth nations, they find that most of the growth differential can be attributed to Nigeria’s higher macro- economic volatility. In turn, simple simulations linking changes in per capita GDP to the level of consumption per capita of households suggest that if Nigeria had had lower levels of volatility and better macroeconomic policies, poverty would have been much lower than observed.
Conclusion
In large part due to lack of data, the documentation on trends in poverty in Africa is limited, and West African countries are no exception to this rule. The objective of this volume has been fairly limited: as part of a broader effort at the World Bank to measure trends in well being in Sub-Saharan Africa, the aim has simply been to provide data on trends in poverty, and on the relationships between poverty, growth and inequality, in six West African coun- tries. While it would not be appropriate to make broad generalizations from the experience of those six countries, the results provide some useful insights.
First, the papers in this volume suggest that long periods of high growth (as in Ghana and Senegal) do lead to poverty reduction, and that when this is apparently not the case (as in Burkina Faso and Cape Verde), the apparent lack of impact of growth on poverty may be due to measurement errors or methodological issues. Second, and at the same time, the impact of growth on poverty has often been more limited than one could have hoped, because inequality has increased in several countries or because growth has not been strictly speaking pro-poor (this is the case in Cape Verde, Ghana, and Senegal). Third, the expe- rience of countries such as Guinea-Bissau and Nigeria suggests on the basis of simple sim- ulations that a lack of growth can be very detrimental for the level of poverty observed in a country, and that conflict and macroeconomic volatility are two key factors that have pre- vented these countries to achieve economic grow in the past.
Overall, the case studies in this volume make a strong case for the positive impact of growth for the reduction in poverty in West Africa, but they also point to the need to pay close attention to changes in inequality as such changes have limited the gains from growth for the poor in several of the countries considered here.
